Assessment of geostatistical features for object-based image classification of contrasted landscape vegetation cover

Accurate mapping and monitoring of savanna and semiarid woodland biomes are needed to support the selection of areas of conservation, to provide sustainable land use, and to improve the understanding of vegetation. The potential of geostatistical features, derived from medium spatial resolution satellite imagery, to characterize contrasted landscape vegetation cover and improve object-based image classification is studied. The study site in Brazil includes cerrado sensu stricto, deciduous forest, and palm swamp vegetation cover. Sentinel 2 and Landsat 8 images were acquired and divided into objects, for each of which a semivariogram was calculated using near-infrared (NIR) and normalized difference vegetation index (NDVI) to extract the set of geostatistical features. The features selected by principal component analysis were used as input data to train a random forest algorithm. Tests were conducted, combining spectral and geostatistical features. Change detection evaluation was performed using a confusion matrix and its accuracies. The semivariogram curves were efficient to characterize spatial heterogeneity, with similar results using NIR and NDVI from Sentinel 2 and Landsat 8. Accuracy was significantly greater when combining geostatistical features with spectral data, suggesting that this method can improve image classification results.

[1]  Qi Chen,et al.  Automatic variogram parameter extraction for textural classification of the panchromatic IKONOS imagery , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[2]  Suha Berberoglu,et al.  Assessing different remote sensing techniques to detect land use/cover changes in the eastern Mediterranean , 2009, Int. J. Appl. Earth Obs. Geoinformation.

[3]  Mario Chica-Olmo,et al.  An assessment of the effectiveness of a random forest classifier for land-cover classification , 2012 .

[4]  Henrique Ferraco Scolforo,et al.  Spatial Distribution of Aboveground Carbon Stock of the Arboreal Vegetation in Brazilian Biomes of Savanna, Atlantic Forest and Semi-Arid Woodland , 2015, PloS one.

[5]  Frieke Van Coillie,et al.  Introduction to the GEOBIA 2010 special issue: From pixels to geographic objects in remote sensing image analysis , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[6]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[7]  Yuhong He,et al.  An object-based approach to delineate wetlands across landscapes of varied disturbance with high spatial resolution satellite imagery ☆ , 2015 .

[8]  Patrick Bogaert,et al.  Forest change detection by statistical object-based method , 2006 .

[9]  Mario Chica-Olmo,et al.  Computing geostatistical image texture for remotely sensed data classification , 2000 .

[10]  Xian Wu,et al.  Evaluation of semivariogram features for object-based image classification , 2015, Geo spatial Inf. Sci..

[11]  P. Curran.  The semivariogram in remote sensing: An introduction , 1988 .

[12]  Txomin Hermosilla,et al.  Change detection in periurban areas based on contextual classification , 2012 .

[13]  C. Woodcock,et al.  The use of variograms in remote sensing. I - Scene models and simulated images. II - Real digital images , 1988 .

[14]  Gang Chen,et al.  International Journal of Applied Earth Observation and Geoinformation Remote Sensing and Object-based Techniques for Mapping Fine-scale Industrial Disturbances , 2022 .

[15]  Luis Ángel Ruiz Fernández,et al.  Definition of a comprehensive set of texture semivariogram features and their evaluation for object-oriented image classification , 2010, Comput. Geosci..

[16]  Chao Zhang,et al.  Texture extraction for object-oriented classification of high spatial resolution remotely sensed images using a semivariogram , 2013 .

[17]  Jan Verbesselt,et al.  Characterizing Forest Change Using Community-Based Monitoring Data and Landsat Time Series , 2016, PloS one.

[18]  Steven E. Franklin,et al.  A comparison of pixel-based and object-based image analysis with selected machine learning algorithms for the classification of agricultural landscapes using SPOT-5 HRG imagery , 2012 .

[19]  Fei Deng,et al.  Integration of orthoimagery and lidar data for object-based urban thematic mapping using random forests , 2013 .

[20]  Thomas Blaschke,et al.  Object based image analysis for remote sensing , 2010 .

[21]  Xuehong Chen,et al.  An improved automated land cover updating approach by integrating with downscaled NDVI time series data , 2015 .

[22]  Joanne C. White,et al.  Optical remotely sensed time series data for land cover classification: A review , 2016 .

[23]  Brian P. Salmon,et al.  Rapid Land Cover Map Updates Using Change Detection and Robust Random Forest Classifiers , 2016, Remote. Sens..

[24]  Henrique Ferraco Scolforo,et al.  Spatial interpolators for improving the mapping of carbon stock of the arboreal vegetation in Brazilian biomes of Atlantic forest and Savanna , 2016 .

[25]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[26]  Luis Ángel Ruiz Fernández,et al.  Using semivariogram indices to analyse heterogeneity in spatial patterns in remotely sensed images , 2013, Comput. Geosci..

[27]  U. Benz,et al.  Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information , 2004 .

[28]  Rick L. Lawrence,et al.  Mapping invasive plants using hyperspectral imagery and Breiman Cutler classifications (RandomForest) , 2006 .

[29]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[30]  Patrick Hostert,et al.  Mapping Brazilian savanna vegetation gradients with Landsat time series , 2016, Int. J. Appl. Earth Obs. Geoinformation.

[31]  Qihao Weng,et al.  A survey of image classification methods and techniques for improving classification performance , 2007 .

[32]  Peter M. Atkinson,et al.  The integration of spectral and textural information using neural networks for land cover mapping in the Mediterranean , 2000 .

[33]  Frédéric Baret,et al.  Multivariate quantification of landscape spatial heterogeneity using variogram models , 2008 .

[34]  Geoffrey J. Hay,et al.  Object-based change detection , 2012 .

[35]  Marjolein F. A. Vogels,et al.  Agricultural cropland mapping using black-and-white aerial photography, Object-Based Image Analysis and Random Forests , 2017, Int. J. Appl. Earth Obs. Geoinformation.

[36]  Mariana Belgiu,et al.  Random forest in remote sensing: A review of applications and future directions , 2016 .

[37]  Txomin Hermosilla,et al.  Description and validation of a new set of object-based temporal geostatistical features for land-use/land-cover change detection , 2016 .

[38]  Koreen Millard,et al.  On the Importance of Training Data Sample Selection in Random Forest Image Classification: A Case Study in Peatland Ecosystem Mapping , 2015, Remote. Sens..

[39]  Ö. Akar,et al.  Integrating multiple texture methods and NDVI to the Random Forest classification algorithm to detect tea and hazelnut plantation areas in northeast Turkey , 2015 .

[40]  Edson E. Sano,et al.  Land cover mapping of the tropical savanna region in Brazil , 2010, Environmental monitoring and assessment.

[41]  R. Mittermeier,et al.  Biodiversity hotspots for conservation priorities , 2000, Nature.

[42]  F. Baret,et al.  Quantifying spatial heterogeneity at the landscape scale using variogram models , 2006 .

[43]  Discriminação da cobertura vegetal do Cerrado matogrossense por meio de imagens MODIS , 2010 .

[44]  Aniruddha Ghosh,et al.  A comparison of selected classification algorithms for mapping bamboo patches in lower Gangetic plains using very high resolution WorldView 2 imagery , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[45]  Txomin Hermosilla,et al.  Original papers: A feature extraction software tool for agricultural object-based image analysis , 2011 .

[46]  Dongmei Chen,et al.  Change detection from remotely sensed images: From pixel-based to object-based approaches , 2013 .

[47]  Freek D. van der Meer,et al.  Remote-sensing image analysis and geostatistics , 2012 .

[48]  Jennifer A. Miller,et al.  Contextual land-cover classification: incorporating spatial dependence in land-cover classification models using random forests and the Getis statistic , 2010 .